Four months after losing partner AstraZeneca plc upon mixed cardiovascular data, AtheroGenics Inc. plans to start looking for a new collaborator, as patient enrollment gets underway in a Phase III trial of its oral anti-inflammatory antioxidant drug, AGI-1067, in diabetes.

The company anticipates beginning discussions relatively soon, likely this fall, said Mark Colonnese, executive vice president and chief financial officer. With the large diabetes market, the drug would be "served well commercially to have a partner," and, he added, the Atlanta-based firm already has seen some interest in AGI-1067, based on "very compelling data" emerging earlier this year from a previous Phase III study in atherosclerosis.

That study, the 6,144-patient Phase III ARISE (Aggressive Reduction of Inflammation Stops Events) showed no difference between AGI-1067 and placebo in the primary endpoint, defined as time to first incident of a composite of major adverse cardiovascular events related to acute coronary syndrome. The news cost AtheroGenics its 2005 whopper of a deal with London-based AstraZeneca, which carried a potential for $650 million in milestones and hefty royalty payment, and prompted to firm to reduce its workforce by half to cut costs. But further analysis of the data showed a clear improvement in glycemic control for diabetic patients receiving AGI-1067, prompting the company in June to shift gears and focus its resources on the diabetes space. (See BioWorld Today, June 1, 2007.)

"There's a great need for additional diabetes therapies that can be added onto" existing drugs, Colonnese said, especially since many patients "are not reaching their standard target blood glucose level" while on those treatments alone. With that in mind, he told BioWorld Today, "we made the strategic change to go into diabetes."

To that end, AtheroGenics began enrolling patients this week in its new Phase III study, titled ANDES (AGI-1067 as a Novel Anti-Diabetic Agent Evaluation Study). Expected to involve about 1,200 patients with Type II diabetes mellitus, the study is designed to test three doses (75 mg, 150 mg and 300 mg) of AGI-1067 vs. placebo over 25 weeks, with the endpoints being glycemic control. The trial is not being conducted under a special protocol assessment, but Colonnese said the "regulatory pathway for [diabetes drugs] is well defined by the agency" and, based on guidance from the FDA, should qualify as a registrational study.

The diabetes indication also offers the company a much shorter path to market. The ANDES trial is expected to conclude before the end of 2008, with an interim analysis planned for the middle of the year.

But that doesn't preclude further development of AGI-1067 in the cardiovascular space. While the previous ARISE trial missed its primary composite endpoint, it did show benefit in the "hard" cardiovascular endpoints, such as death, heart attack and stroke, Colonnese said. "That's part of our long-term view of the drug," though a cardiovascular trial likely would be longer and more expensive than the diabetes study.

For now, the company is positioned well to complete Phase III testing of AGI-1067 in diabetes, with or without a partner. AtheroGenics, which reported a net loss of $6.1 million, or 16 cents per share, for the second quarter, had $115 million in cash as of June 30. The company also reduced its near-term cash requirements last month, with the exchange of $38 million of convertible debt due in 2008 for $60 million in convertible debt due 2011.

"We have all the money we need for this trial," Colonnese said.

Shares of AtheroGenics (NASDAQ:AGIX) gained 11 cents Thursday to close at $1.44.